Inflation-adjusted median incomes in Colorado have moved little over the past decade

Now that median household income data is available for 2014, let's take a new look at how incomes are doing in Colorado.

There are numerous measures of median incomes, but two of the easiest to find and most widely used  are the 5-year estimates from the American Community Survey (ACS) and the data from the Annual Social and Economic Supplement (CPS ASEC) from the Census Bureau.

Let's begin with the ACS data.

Using 5-year estimates in current dollars from the ACS, we get the following values for median household income:

According to the ACS, the 5-year estimate for median household income in 2014 was $59,448. That was up 1.7 percent from the year before:

Overall, however, we can see that over the past four years, the median income has been rather stable.

If we adjust for inflation, however, we find that the median incomes have been declining slightly:

There are numerous implications to this, of course. If median incomes are basically flat, but home prices are increasing at a rate of five to ten percent, this will certainly impact the affordability of housing. We'll look at this in future posts.

Other Census Data 

Using the 1-year median income measure from the Annual Social and Economic Supplement (CPS ASEC), we get the following:

Here we find a bit more volatility in the numbers, although, like the ACS data, they tend to hover around $60,000.

Measured in terms of year-over-year changes, we find that 2014 saw an decrease of 3.8 percent in its median income, which followed a very large increase for 2013 over 2012:

Overall, we can conclude that median income hasn't really moved much since 2007 when it peaked toward the end of the last economic expansion. In 2013, median incomes hit a new peak, but then fell below 2007 levels again in 2014.

The picture sours a bit more when we adjust these values for inflation. In 2014 dollars:

In this case, it's more clear that Colorado's median income has not returned to where it was in 2007. More or less, Colorado's median income is today where it was a decade ago.

Regardless of which measure we use here, it's likely safe to say that the median income in Colorado is somewhere around $60,000. Since the end of the last recession in 2009, nominal incomes have increased somewhat, although it is unclear if, once we adjust for inflation, whether or not households have made much headway relative to where we were at the peak of the last economic expansion.

*To adjust for inflation, I've used the Denver-Boulder-Greeley CPI. All graphs in this article are for Colorado statewide.

Are more twenty-somethings living at home in Colorado?

Household formation has long been an issue central to the demand for real estate. If people move out of their parents homes and create a new household, then a new housing unit will be demanded. If two people move out, and get one unit together then one new unit will be created out of two. If both people can afford to get their own apartments, then two new units will be created out of two.

Economic prosperity has long been connected to economic prosperity. If incomes are low, or housing costs are high, people will either stay at home or take on additional roommates to afford housing. If wages are high or housing costs are low,  more people will demand more units. This is moderated, of course, by people cohabiting for romantic/family reasons, such as marriage. In that case, two households will reduce to one even when economic times are good. 

Nevertheless, on the whole, there is reason to believe that when incomes and economic prosperity increase,  people tend to demand more housing units.

Are Young People Now Too Poor to Move Out? 

Last year, the New York Fed published an analysis on how many 25-year olds were living with their parents.  Here are their results: 

In 2003, between 20 and 30 percent of twenty-five-year-olds lived with their parents (using our measure) in twenty-five of the forty-eight states. By 2013, all forty-eight states had parental co-residence rates of more than 30 percent. Indeed, for twelve states, the parental co-residence rate for twenty-five-year-olds had risen above 50 percent. Four states—Maine, Minnesota, New Hampshire, and Vermont—saw the rate at which twenty-five-year-olds live with their parents increase by more than twenty percentage points between 2003 and 2013. Parental co-residence was highest in Mid-Atlantic and Southern states in 2003, but by 2013 it was highest in the Northeast and Midwest. 

So, for the period of 2003-2013, there was indeed an increase in the number of people living at home. Here's what it looked like in 2003.  Colorado is in the 20%-30% range: 

But, by 2013, here's what it looked like. Colorado is in the 30%-40% range: 

In both cases, Colorado is ranked among the states with the fewest 25-year olds living at home. 

The NY Fed report goes on: 

Parental co-residence increased steadily for both age groups from at least 2003 through 2012, followed by a leveling off or slight decline in 2013. The chart also shows one measurement of household formation—homeownership—which has been decreasing for both twenty-five- and thirty-year-olds since 2007, the end of the housing bubble and the start of the Great Recession. While thirty-year-olds were twice as likely to own a home as they were to live with their parents in 2003, we find that they were equally likely to own a home or live with their parents in 2013. 

So what are the reasons for this? The report attempted to address that too: 

Our results demonstrate that local economic growth is a mixed blessing when it comes to building youth independence: Improvement in youth employment conditions enables young people to move away from their parents, but rising local house prices are estimated to have forced many young people to move back home. These two effects partially offset each other. 
However, the relationship we observe between rising student debt and co-residence with parents is clearer. The chart below presents a state-level scatter plot of the change in the rate of living with parents from 2008 to 2013 against the change in average student debt per graduate. 

It reveals a clear positive correlation between a state’s student debt growth and the rate at which its twenty-five-year-olds live with their parents. The regression line in the chart indicates that a $10,000 increase in student debt per graduate in the state is associated with an additional 2.9 percentage point rise in the rate of living with parents. (Estimates in the staff report that account for changes in the local economy and other factors tell a similar story.) 

So how does Colorado compare in terms of student debt? Fortunately for us, the Dallas Fed released a 2014 report on this, and the map looks like this: 

In Colorado, the mean (average) balance was $26,215, which puts it at 16th highest nationwide. 

Based on this statistic alone, then, we'd expect Colorado to have high rates of people living at home. But that's not the case. Colorado has  some of the lowest rates of people living at home. As a possible explanation, we might look to the fact that that Colorado has the 12th highest median income among the states. 

According to Census data, Colorado household median income was $60,940 in 2014, which put it above the national median household income of $53,657. (The highest state median income was found in Maryland at $76,165.)

Colorado may have relatively high student debt, but it's incomes may be  factor in making up for that. Moreover, in this case we're looking at average student debt and median incomes. The median incomes suggest that the incomes reflect a relatively typical income level.  It's why we often prefer the median over the average. But, the student debt level here is an average which means it could be skewed  up by a small number of people with very large debt levels. From this we might conclude it is indeed plausible that, at least in the case of Colorado, student is not the dominating factor in the growth of living at home. 

Related post: "A Better View of Poverty Rates: We Must Consider the Cost of Living." 

Federal spending, state by state

The Pew Charitable trusts recently released new data up through 2013 on federal spending in the states.

If we look at federal spending as a proportion of each state's overall GDP, we find that the recipients are not exactly evenly distributed:

Source: Pew Charitable Trusts  (based on data from 2004–2013)

This is all federal spending, so these totals are a combination of military spending, social welfare programs such as Medicare, and ordinary civilian federal spending, including civilian research facilities and other programs funded by federal grants.

These are proportional numbers, so they are a function of both the amount of federal spending as well as the overall size of GDP. So, in California, for example, which receives immense amounts of government spending, is nevertheless a state where federal spending is offset by a very large private sector. In a more rural state with few major private industries, such as New Mexico, the state shows up as being highly reliant on federal spending.

By this measure, the state most reliant on federal spending is Mississippi where federal spending is equal to 32 percent of the state's GDP. The state least reliant on federal spending is Wyoming where federal spending is equal to 11 percent of the state's GDP:

Source: Pew Charitable Trusts  (based on data from 2004–2013)

The above measure gives us a sense of how much federal spending is taking place relative to overall economic activity. But, it tells us little about how much the feds are spending in each state relative to the tax revenue being produced in each state.

To discover that, we need to compare federal spending to tax collections from each state. So, I took gross tax collection by state, and then subtracted refund totals. I then compared the "net" collections to Pew's total federal spending data in each state. (The tax data used was 2013 data.) We can then measure the result in terms of dollars spend in each state per dollar in tax revenue collected. States that have a value of less than a dollar in the map below receive less than a dollar in federal spending for every dollar in taxes paid from that state. So, for example, Ohio receives 91 cents in federal spending for every dollar collected in taxes from Ohio:

I've divided this graph up into "net tax payer states," "break-even states" and "net tax receiver" states. The lightest shade of blue are states that, by far, pay in more than they receive back, such as New Jersey and Minnesota. The next lightest shade of blue are states that are more or less "break even" in the sense that spending and tax collections hover somewhat around a 1-for-1 relationship. The darker blue states are states that receive considerably more in federal spending than they pay in taxes.
Here are all states, including values:

Naturally, these values aren't spread evenly within the states themselves, either. Areas that are more rural and reliant on agriculture will tend to be net tax receiver areas both because farmers and ranchers receive a lot of government subsidies, and also because agricultural work tends to have lower productivity than urban work.

Urban areas, in contrast, produce most of the tax revenue, so highly urbanized states will tend to more often be "break even" or "net tax payer" states.

Other Considerations

One thing that must not be ignored is the fact that the US government spends more than it takes in nationwide. During 2013, for example, the federal government spent a dollar for every 80 cents it took in via taxes.

Nationwide, the tax-spending ratio is not one dollar, but it about $1.20. So, states that are getting around $1.20 back for every dollar extracted in taxes are really just at the national average.
This is being made possible by old-fashioned deficit spending and also by monetization of the debt which the Federal Reserve facilitates by expanding the money supply. Once interest rates rise or the international value of the dollar begins to fall significantly, this sort of overspending will no longer be possible, and many states will find themselves in dire straits. (States that are "net tax payer" or "break even" states will adjust the best to any significant disruptions in federal spending.)

A Better View of Poverty Rates: We Must Consider the Cost of Living

This week a number of wire services picked up a story in which states are ranked according to which states are the "most expensive states to raise a family." The list, which was created by a private company to drive web traffic to its site, attempts to quantify the cost of raising a family by factoring in government mandated family leave, the cost of child care, and other factors.

The use of mandatory family leave is rather novel, given that mandated leave raises the effective minimum wage for many workers, and thus negatively impacts the least-skilled workers the most.  Nevertheless, the list appealed to the common-sense notion that there's more to one's standard of living than a relatively high income. The cost of living is an important factor.

The US Poverty Rate Does Not Account for Local Cost of Living 

Given the importance of the cost of living, it is very problematic that the official poverty rate totals for US states do not take costs into account.

When measuring poverty rates internationally, poverty is just defined as households that make 50 percent or 60 percent of the national median income. Although these measures often attempt to take into account differences in the cost of living among different countries, measuring poverty this way provides its own set of problems. It simply makes poverty a purely relative measure, so we end up with a situation where purchasing power for a median household in one country (say, Portugal) is actually lower than a poverty-level household in another country (say, the US).

The US official measure, on the other hand, attempts to get around this problem by defining the poverty rate as an actual dollar amount based on what a household can buy. The federal government has set the poverty income at $24,250 for a family of four in 2015.

The problem is this dollar amount is applied nationwide and then used to calculate poverty rates. So, a household in Arkansas at this income level is deemed "poor" while a household in California at the same income level is deemed equally poor. However, the cost of living in much of Arkansas is quite a bit lower than in much of California.

If we fail to adjust for the cost of living, the poverty rate  map looks like this:

In this case, the highest poverty rate is found in Mississippi with a rate of 23 percent, with Arizona and New Mexico close behind at 21 percent and 19 percent, respectively. New York and California are a dozen states down the list with poverty rates 15 percent for both. (See here for full list based on 2009 calculations.)

Many have noticed certain regional trends here, and that has led to a myriad of articles claiming that so-called "red states" have higher poverty rates than the "blue states." In many cases, "red states" is really code for "low tax" or "free-market-ish" state. In other words, this map "proves" that low taxes and freer economies cause more poverty.

This might be a conundrum if it were not for the fact that this measure of poverty completely ignores the plight of low-income households in states where the cost of living is very high. The biggest offenders here are, not surprisingly, California and New York, where rents and the cost of living in general is very high.

The feds have long recognized the discrepancy here, and in the fine print have long noted that poverty rates should only be used as very general "guidelines" or measures over time. Comparisons among states are discouraged.

That doesn't stop pundits from claiming that blue states like California and New York have been successful in combating poverty through tighter regulation of business, and higher taxation.

If we adjust the states and poverty rates for the cost of living, however, the map looks a bit different:

In this case, the state with the highest poverty rate is California at  23 percent. Arizona and Florida are close behind with rates of 22 percent and 20 percent, respectively. New York has risen to sixth place with a poverty rate of 18 percent, while Mississippi has fallen to eighth place with a rate of 17 percent.

Here we see our bias-confirming assumptions no longer seem to apply since  no correlation is apparent along the lines of the red-state/blue-state claims. Right-wing Indiana, at 15 percent, is more or less equal with left-wing Illinois, while Mississippi and New York, with widely divergent public policy regimes, also have similar poverty rates. (See Table 3.)

Obviously, we have to look somewhere beyond our neat-and-nice ideas about red states and blue states to come up with an explanation.

Of course, poverty can be a function of so many things, that it's impossible to generalize. Public policy is certainly a factor, but so are cultural factors, access to capital, the transportation infrastructure, and more. Some states are influenced by the presence of Indian Reservations (such as Arizona) where local economic policy is more influenced by federal law than state law.

But most of the discussion about "rich states" and "poor states" has long been skewed by the fact we tend to ignore cost of living.

As a final illustration, we can look at median incomes from each state. The median income figures put out by the census bureau do not account for regional "price parity."

Using just the basic median income numbers from the Current Population Survey, we get this:

The US median income is $51,849, and many high-cost states come in well above this, with Hawaii at $59,244 and California at $57,688.  Meanwhile, Mississippi and Louisiana come in at $39,011 and $39,637, respectively.

State median incomes vary by as much as $31,000, with New Hampshire coming in at $70,063 which is $31,051 higher than Mississippi.

But once we adjust incomes for price parity, we find that many of the high-income, high-cost states fall quite a bit in the list:

First, we notice that this compresses the variation in median incomes. The difference between the highest-income state (New Hampshire at $66,159) and the lowest income state (Louisiana at $43,462), shrinks to $22,697.

We also notice that New York now has the second-lowest median income in the country, right between Louisiana and Mississippi. New York now has a real median income of $44,326, while Mississippi has a real median income of $44,944. California and Hawaii fall from being near the top of the list to below the national median income, with median incomes of $51,369 in California and $50,984 in Hawaii.

Basically, the purchasing power of a median household in New York or California is much lower than has been traditionally suggested.

This is also important to keep in mind when comparing US median incomes and poverty rates to foreign countries. Much of the US is very inexpensive in terms of cost of living and well below northern Europe, New Zealand, Australia, and even Canada.
Maps and graphs by Ryan McMaken.

Is Living-at-Home an Indicator of the Standard of Living?

In this article, I noted that median incomes — even when adjusted for cost of living, taxes, and social benfits — are higher in the United States than in Europe. What's more, Americans at poverty level (i.e., 60% of national median income) have more purchasing power than median-level households in many European countries.

But what are some outward indications of this? The UNICEF report on childhood poverty, for example, attempts to quantify the effects of poverty by asking extremely subjective questions like "Did you feel stress this week?" and using the answer to compile a type of index. The problems with indicators such as these should be obvious, since "stress" can mean any number of things.

But let's try for a somewhat more concrete manifestation of a low standard of living: adults living with parents.

In the US, at least, it's been long accepted that new household formation is an indicator of the state of the economy. Specifically, it is assumed that young people who can find gainful employment will be able to move out of their parents' homes more quickly, and also potentially find spouses and/or have children sooner. Economic factors are generally blamed on trends in living at home, as with the discussion about the "boomerang effect" during the most recent business cycle.

This has been the case historically in the US as the marrying age of Americans has fluctuated over time and geography with the availability of land, which was a key factor in economic self-sufficiency in an agrarian setting.

It appears to have been an important factor in Europe as well, since in pre-industrial times, there was little land available for purchase or settlement — and even less in the way of manufacturing jobs — and adult children often had to wait to simply inherent the family lands rather than attempt to find a new household.

So what percentage of adults ages 25-34 still live with their parents nowadays? According to Eurostat, the percentage of adults in this group that still live at home varies substantially from country to country:

The US numbers are from the Census. (Men tend to live with their parents in much larger percentages than women, by the way.)

The percentages here swing quite a bit between 57 percent in Slovakia and Denmark at 1.4 percent.

It's reasonable to accept that cultural factors may be at play here.  Clearly, living at home with one's parents appears to be frowned upon more in Scandinavia than in other rich countries of similar median income levels.

Nevertheless, if we do plot the living-home numbers against median disposable income, we do see a pretty clear correlation:

The countries where the median household has more purchasing power have fewer cases of adult children living with parents. Few should be surprised to find Greece and Portugal up in the top left of graph, for example. Almost all countries with median incomes above $20,000 have living-at-home percentages below 20 percent, while almost all countries with median income below $30,000 have living-at-home levels above 30 percent.

While it's not a perfect proxy for household purchasing power, the percentage of adults that continue to live at home in the prime family formation years does give us some insight into the real-world implications of the fact that real disposable income is lower in most European countries than it is in the United States.

A note on the data: This is all 2013 data, except for Turkey, which is 2007 (the most recent available.) The US data matches up with the Euro data on the age of the adults measured (i.e., 25-34), although in the case of the Euro data, it includes married or cohabiting adults children living with parents. These people are excluded in the US data. Fortunately for us, though, the percentage of people living at home who were also married or cohabiting is under 5 percent. 

A Quick Look at Median Household Income Up to 2013

The Census Bureau won't release 2014's median household income for another month or so, but we can have a look at trends up through 2014, for now. These numbers are not adjusted for inflation.

For 2013, the median household income in Colorado was $63,371. In the US for the same period, it was $51,939.

As we can see in the first graph, the Colorado median income level has been above the US level since 1990.

 

We can also see that over time, this gap has been growing. The second graph shows the gap between the Colorado median income and the US median income: 

 

The gap was negative from 1986 to 1989 when Colorado's median income was lower than the US. But since then, the gap has generally grown, and reached $10,000 for the first time in 2007. In fact, 2013's gap was the largest ever recorded with Colorado's median income coming in at 11,432 above the US level. 

The final graph shows percentage change in median income for each area. Colorado's YOY changes are much more volatile, as would be expected from an area so much smaller than the US overall. The sheer size of the US and its economy prevent large swings. However, there is a slight downward drift in the US median income increases over time. In other words, US median income seems to be going up by a smaller amount over time. In Colorado, what was a downward drift during the 1990s, appears to have stabilized somewhat since 2003. The YOY change in Colorado from 2012 to 2013 (10.6 percent) was the second largest ever recorded, second only to 1990's growth rate of 14.6 percent. The US rate of change for 2013 was 1.8 percent.